hushem_1x_deit_small_sgd_00001_fold3

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.5939
• Accuracy: 0.1860

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	6	1.6009	0.1860
1.5096	2.0	12	1.6005	0.1860
1.5096	3.0	18	1.6002	0.1860
1.4896	4.0	24	1.5998	0.1860
1.4946	5.0	30	1.5995	0.1860
1.4946	6.0	36	1.5992	0.1860
1.508	7.0	42	1.5988	0.1860
1.508	8.0	48	1.5986	0.1860
1.4945	9.0	54	1.5983	0.1860
1.5205	10.0	60	1.5980	0.1860
1.5205	11.0	66	1.5977	0.1860
1.5058	12.0	72	1.5975	0.1860
1.5058	13.0	78	1.5972	0.1860
1.5082	14.0	84	1.5970	0.1860
1.502	15.0	90	1.5967	0.1860
1.502	16.0	96	1.5965	0.1860
1.5281	17.0	102	1.5963	0.1860
1.5281	18.0	108	1.5961	0.1860
1.4713	19.0	114	1.5959	0.1860
1.5067	20.0	120	1.5957	0.1860
1.5067	21.0	126	1.5955	0.1860
1.5046	22.0	132	1.5953	0.1860
1.5046	23.0	138	1.5952	0.1860
1.4884	24.0	144	1.5950	0.1860
1.4923	25.0	150	1.5949	0.1860
1.4923	26.0	156	1.5948	0.1860
1.4973	27.0	162	1.5947	0.1860
1.4973	28.0	168	1.5945	0.1860
1.5002	29.0	174	1.5945	0.1860
1.4807	30.0	180	1.5944	0.1860
1.4807	31.0	186	1.5943	0.1860
1.486	32.0	192	1.5942	0.1860
1.486	33.0	198	1.5941	0.1860
1.4927	34.0	204	1.5941	0.1860
1.4875	35.0	210	1.5940	0.1860
1.4875	36.0	216	1.5940	0.1860
1.5166	37.0	222	1.5940	0.1860
1.5166	38.0	228	1.5939	0.1860
1.5127	39.0	234	1.5939	0.1860
1.4974	40.0	240	1.5939	0.1860
1.4974	41.0	246	1.5939	0.1860
1.4716	42.0	252	1.5939	0.1860
1.4716	43.0	258	1.5939	0.1860
1.5277	44.0	264	1.5939	0.1860
1.501	45.0	270	1.5939	0.1860
1.501	46.0	276	1.5939	0.1860
1.4805	47.0	282	1.5939	0.1860
1.4805	48.0	288	1.5939	0.1860
1.5052	49.0	294	1.5939	0.1860
1.536	50.0	300	1.5939	0.1860

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1